AWARE SYSTEMS
TIFF and LibTiff Mail List Archive

Thread

1998.11.12 15:00 "Revison 7.0", by Dr. Klaus Bartz
1998.11.12 15:54 "Re: Revison 7.0", by Tom Lane
1998.11.13 15:50 "Re: Revison 7.0", by Dr. Klaus Bartz
1998.11.13 15:59 "Re: Revison 7.0", by Tom Lane
1998.11.12 22:38 "Re: Revison 7.0", by Mark Knibbs

1998.11.12 15:54 "Re: Revison 7.0", by Tom Lane

Is jbig dead? ( I know, hard coding, some patent ). Two weeks ago http://www.jbig.org/ was dead and in this moment an open with this URL shows an other home site

Try www.jpeg.org --- same ISO committee. (Right at the moment visiting www.jbig.org redirects my browser to www.jpeg.org. Dunno why it's not working for you.)

But, as you say, JBIG is no help if you want a patent-free solution.

Has some one tested bzip2?? I have seen dramatical results at a 4 color image imported from HPGL/2. TIFF RAW 79 MB, TIFF deflate 950 KB, TIFF RAW and then bzip2 290 KB. Very special, yes.

FWIW, the PNG group looked at what the difference would be if we substituted bzip2 for deflate in PNG. We saw numbers like 10-30%, IIRC --- definitely less than a factor of two anyway. So I think you have a very unusual case here.

The PNG group found that a differencing filter in front of the deflate engine helps a lot for some kinds of images. If there's going to be any serious work on inserting deflate into the TIFF spec, it needs to go along with some extension of the Predictor tag. I imagine the same would be true if bzip2 were picked instead.

                        regards, tom lane
                        organizer, Independent JPEG Group